Prof. Xiaohua Huang, Nanjing Institute of Technology, China

Xiaohua Huang is Professor at Nanjing Institute of Technology, Ph.D., Master's Supervisor, IEEE/CSIG/CCF Senior Member, Distinguished Professor of Jiangsu Province, High-level Talent of Jiangsu "333 Project" (Second Level), and has been listed among the world's top 2% scientists for consecutive years. He has led or participated in over 10 domestic and international projects funded by the National Natural Science Foundation of China (General Program), the Natural Science Foundation of Jiangsu Province (General Program and Youth Program), the Academy of Finland, Business Finland, the Nokia Foundation, and others. Has published over 90 papers in domestic and international journals and conferences, including IEEE Transactions, and has received multiple awards such as the Second Prize of the Natural Science Award of the Ministry of Education and the Second Prize of the Beijing Science and Technology (Natural Science) Award. His research interesting include affective perception and interaction, computer vision and industrial visual inspection, embodied intelligence, and multimodal perception.

 

Speech Title: Emotion Recognition for Intelligent Service Robots: A Study on Technological Exploration and Application Prospects

Abstract: This invited talk introduces the frontier topic of emotion recognition for intelligent service robots. Starting from national strategic needs, it highlights that emotion-intelligence integration is an inevitable trend for intelligent service robots and reviews policy support for affective interaction technologies. Based on psychological research, the talk introduces the basic theories and practical challenges of facial expression, micro-expression, and group-level emotion recognition. Building on our research group's accumulated work in affective computing, it also shares our progress in facial expression recognition, micro-expression analysis, and group emotion perception. Finally, the talk outlines future directions such as multimodal sentiment analysis, emotionally intelligent digital humans and robots, and affective perception in complex scenarios, while highlighting ethical and societal considerations for real-world deployment.

Prof. Heming Jia, Sanming University, China

Heming Jia, Professor, PhD in Systems Engineering from Harbin Engineering University, Master’s Supervisor, Expert of Industrial Metrology in Fujian Province. Academic Leader of the Electronic Information Discipline at Sanming University. He was selected for the 2026 World’s Top 2% Scientists List for Lifetime Scientific Impact, and consecutively included in the World’s Top 2% Scientists List for Annual Scientific Impact from 2023 to 2026. He is also among the Top 1% Highly Cited Scholars of CNKI 2024. He is an Outstanding Teacher of Sanming City, Class C High-level Talent of Sanming City, Executive Director of Fujian Artificial Intelligence Society, Director of Fujian Computer Society, Member of the National University Electronic Information Discipline Construction Committee of the Chinese Institute of Electronics, Member of the Youth Working Committee of the Chinese Association for Artificial Intelligence, and Senior Member of the China Computer Federation. He serves as an Editorial Board Member of the SCI-indexed journal *Biomimetics*, an Editorial Board Member of the EI-indexed journal *Information*, an Assistant Editorial Board Member of the Chinese core journal *CAAI Transactions on Intelligent Systems*, and a Youth Editorial Board Member of the Chinese core journal *Journal of Computer Research and Development*. He is the recipient of the Golden Phoenix Young Scholar Award and the “Teacher in My Heart” honor at Sanming University. His main research interests lie in intelligent optimization algorithms and their applications. He has published more than 100 research papers as the first author and corresponding author, including over 70 SCI-indexed papers, among which 7 are ESI highly cited papers. He has proposed original intelligent optimization algorithms: the Remora Optimization Algorithm (ROA), the Crayfish Optimization Algorithm (CFOA), the Catch Fish Optimization Algorithm (CFOA), and the Superb Fairy-wren Optimization Algorithm (SFWOA).

 

Speech Title: Novel Swarm Intelligence Optimizers: Design, Mechanism and Applications

Abstract: Swarm‑Intelligence Optimization Algorithms (SIOAs), as a class of stochastic meta‑heuristic computational intelligence approaches, mimic collective collaborative behaviors of natural organisms to solve complex optimization problems. Compared with traditional gradient‑based optimization techniques, SIOAs exhibit prominent strengths including strong global search capability, independence from initial conditions and high robustness, yet they still suffer from limitations such as slow convergence and insufficient local exploitation. This report first introduces the basic concepts, mechanisms, advantages and drawbacks of swarm‑intelligence optimization algorithms. Then, several original bio‑inspired swarm optimizers proposed by our group are elaborated, including the Remora Optimization Algorithm (ROA), Crayfish Optimization Algorithm (COA), Catch‑Fish Optimization Algorithm (CFOA), and Superb Fairy‑wren Optimization Algorithm (SFWOA). Their biological inspiration, mathematical models, search mechanisms and algorithm workflows are systematically illustrated. Furthermore, representative engineering applications of these algorithms are demonstrated, taking hyperspectral image band‑selection as a typical case. By adopting cubic chaotic mapping, sine‑cosine strategy and grouping recommendation mechanism, improved variants of ROA achieve effective dimensionality reduction for hyperspectral data. Finally, open challenges and future research directions are discussed, covering algorithm hybridization, performance‑boosting strategy design and development of novel nature‑driven meta‑heuristics.

Dr. Julia Sidorova, National Consortium for Biomedical Research (CIBER), National Institute of Health Carlos III, Spain

Dr. Julia Sidorova is a Senior Researcher (R3) at the National Consortium for Biomedical Research (CIBER), National Institute of Health Carlos III. Her PhD was awarded by Universidad Pompeu Fabra (2009) for international research (incl. Stanford University and Saarland University) for the construction of one of first five emotion recognizers from voice and its applcation in neurology. After being a postdoctoral grant holder for 4 years (incl. at Universidad Carlos III de Madrid and ETH-Zurich), as an Assistant Professor and a Senior IEEE member she engaged in industrial Big Data research (incl. Sony, Telenor, and Ericsson). She was a research lead in one of the first projects in blood glucose approximation from voice. Her Teaching Philosophy was shaped by innovating teaching statistics, the creation of new university majors at Universidad Carlos III de Madrid, and teaching deep learning to the industrial Master’s at KTH Royal Institute of Technology (>20 theses), Stockholm.  For the past five years, she worked for the National Consortium of Hospitals defending clinical hypotheses in pursuit for more accessible and affordable biomarkers, which resulted in tangible improvements in clinical diagnostics and treatments. Her interests lie in classical data analysis for multi- omics data vs deep neural networks, -- understanding their suitability or deficiencies. She is an author of >30 articles in central journals.

 

Speech Title: Optimization strategies in cleaning the biomedical data from unwanted biological and technical variability

Abstract: The motivation is to correct the experimental design whether the failure in data collection has become known, or remains to be unknown.  The talk will summarize the state of the art in signal cleaning with good practices used in the bioinformatics and engineering communities. The novel algorithm for improved data cleaning will be presented.